SUPPLEMENTARY INFORMATION

Size: px

Start display at page:

Download "SUPPLEMENTARY INFORMATION"

Jean Bradford
6 years ago
Views:

cell types & differentiated cells (Fig 2) Analysis of transcriptional somatic cell memory in each type of hipsc (Fig 3) Correlation between differential gene expression and DNA methylation/bisulfite

1 DOI: /ncb2239 Hepatocytes (Endoderm) Foreskin fibroblasts (Mesoderm) Melanocytes (Ectoderm) +Dox rtta TetO CMV O,S,M,K & N H1 hes H7 hes H9 hes H1 hes-derived EBs H7 hes-derived EBs H9 hes-derived EBs Affymetrix Human Gene ST 1.0 arrays Pluripotency validation for the Hepatocyte-iPS cells derived and used for the microarray analysis (Fig 1) Comparison of transcriptional profiles between hescs, hipscs derived from different cell types & differentiated cells (Fig 2) Analysis of transcriptional somatic cell memory in each type of hipsc (Fig 3) Correlation between differential gene expression and DNA methylation/bisulfite sequencing analysis (Fig 4) Meta-analysis of DNA methylation-associated transcriptional memory in independent datasets (Fig 5) Functional analysis of the somatic cell memory gene C9orf64 during reprogramming (Fig 6) Proximity in the genome affects efficiency of gene silencing in ips cells (Fig 7) Model for the role of DNA methylation in differences between human ips cells and ES cells (Fig 8) Figure S1 Flowchart of analysis and data reported in the manuscript. Human somatic cells representative of the 3 embryonic germ layers were reprogrammed to pluripotency using the dox-inducible lentiviral transgene system for overexpression of the reprogramming factors OCT4 (O), SOX2 (S), cmyc (M), KLF4 (K) and NANOG (N). Hepatocyte-derived ips cells, newborn foreskin fibroblast-derived ips cells and melanocyte-derived ips cells represented the endodermal, mesodermal and ectodermal lineages, respectively. All 3 types of ips cells, their parental somatic cell counterparts, 3 lines of human ES cells (H1, H7, H9) and Embryoid Bodies derived from these ES cells were hybridized to Affymetrix Human Gene ST 1.0 arrays and transcriptionally profiled in parallel. Triplicates of independent clones were used for all cell types except for the somatic cells, since each somatic cell represents a single clone. For all 3 types of parental somatic cells, technical triplicates were used for the analysis. Shown in the box beneath the flowchart is a brief description of each of the figures presented in the main text of the manuscript. 1

2 Figure S2 Pluripotency validation for the Fibroblast-iPS cells derived and used for the microarray studies. (a) All 3 Fib-iPS clones used in this analysis showed strong, positive staining for the human ES cell specific-markers SSEA-3, SSEA-4, Tra1-60 and Tra1-81, comparable to that observed in control H9 ES cells. Scale bar represents 300 μm. (b) qrt-pcr was used to confirm both high expression levels of endogenous pluripotency genes in all 3 Fib-iPS cell clones, as well as negligible levels of transgene expression. Values were standardized to GAPDH and Ubb, then normalized to H9 ES cells (endogenous) or 5-factor infected BJ fibroblasts + dox for 4 days (viral). (c) All clones of Fib-iPS cells formed embryoid bodies in vitro when grown under non-attachment conditions. Shown here are d8 EBs alongside with control ES cell-derived EBs. Scale bar represents 200 μm. (d) The pluripotent nature of all ips cells used in our analysis was confirmed by their ability to form EBs comprised of tissues derivative of the 3 germ layers in vitro (also see Fig. 1c and Supplementary Figure 2c). qrt-pcr analysis on d8 EBs derived from all types of ips cells and ES cell controls confirmed the presence of the 3 embryonic germ layers. Values were standardized to GAPDH and Ubb. Expression fold changes shown in the graph are relative to H9 ES cells on d0. (e) Hematoxylin and eosin stain of teratomas generated from fibroblastderived ips cells injected subcutaneously into immunocompromised SCID mice. Structures derivative of all three germ layers could be identified. (i) Neural tissue (ectoderm), (ii) Striated muscle and mesenchyme (mesoderm), (iii) gut-like epithelium (endoderm). In b and d, data are from triplicate reactions and error bars represent standard deviations. 2

3 a b Figure S3 ips cells retain a transcriptional memory of the original somatic cell. (a) Similar to Fig. 3a, these figures show LOESS curves fitted to the scatter plots of t-test log p-values for fibroblast and melanocyte: -log(p) and log(p) are plotted for fold changes greater than 1 and less than 1, respectively. The black line is a curve fitted to our data, and other curves are fitted to the 1000 bootstrap simulation datasets obtained by assuming identically distributed ips and ES cell expression levels. The black line shows clear deviation from the null hypothesis ips=es and thus reflects the trend that the transcriptional memory of the originating cell type is retained in ips cells: genes that were higher (or lower) in the somatic cell than in ES cells tend to be significantly repressed (or induced) during reprogramming, but nevertheless remain higher (or lower) in ips cells than in ES cells. (b) The partition of differentially expressed genes between ips cells and ES cells according to their expression status in somatic cells. It is seen that ~50% of the genes were already differentially expressed between the corresponding somatic cell type and ES cells, while ~10% were differentially expressed only in the corresponding somatic cell type and not other cell types relative to ES cells. Cell type-specific somatic expression can thus explain ~10% of the observed incomplete reprogramming. 3

b c Figure S4 Targeted differentiation of ips and ES cells towards an endodermal fate. (a) A targeted differentiation approach based on a previously published protocol by Kroon et al.

stage (d6 of the protocol) using the growth conditions shown.

SOX7 (extraembryonic endoderm-specific marker) expression was used to monitor the formation of definitive endoderm since SOX17, FOXA2 and CXCR4 can be found in all endodermal tissues.

4 b c Figure S4 Targeted differentiation of ips and ES cells towards an endodermal fate. (a) A targeted differentiation approach based on a previously published protocol by Kroon et al., 2008 was used to differentiate 2 clones each of Fib-iPS and Hep-iPS cells and 2 lines of ES cells (H1 and H9) to the definitive endoderm stage (d3 of the protocol) and to the primitive gut tube stage (d6 of the protocol) using the growth conditions shown. (b) On d3 of the differentiation assay, qrt-pcr was used to analyze the expression levels of 3 definitive endoderm markers (SOX17, FOXA2, CXCR4) in each of the cell types analyzed. SOX7 (extraembryonic endoderm-specific marker) expression was used to monitor the formation of definitive endoderm since SOX17, FOXA2 and CXCR4 can be found in all endodermal tissues. Low SOX7 levels indicated that the tissue generated was definitive endoderm and low OCT4 expression levels indicated the loss of pluripotency in ips and ES cells during differentiation. (c) On d6, the expression level of 2 primitive gut tube markers, HNF1B and HNF4A, was analyzed by qrt-pcr. The same controls as on d3 were used in this analysis. In both graphs, values were standardized to Ubb and TBP, and the expression fold changes for all genes in each cell type are referenced to a corresponding d0 sample. In b and c, data are from triplicate reactions and error bars represent standard deviations. 4

(a) On d3 of the targeted differentiation protocol, the same 2 clones of Fib-iPS (BJ 1 and 2) and Hep-iPS (Hep 1 and 2) cells, as well as 2 human ES cell lines (H1 and H9), analyzed by qrt-pcr in

5 a D3 - Definitive endoderm stage: b D6 - Primitive gut tube stage: 58.4% 76.0% 92.3% 86.0% 72.7% 87.7% 72.1% 76.3% 83.9% 84.2% 76.4% 93.1% 77.2% 76.7% 51.8% 69.9% 34.8% 44.9% 43.6% 51.9% 66.2% 52.8% 57.8% 56.8% Figure S5 Immunohistochemical analysis of ips and ES cells differentiated towards endoderm. (a) On d3 of the targeted differentiation protocol, the same 2 clones of Fib-iPS (BJ 1 and 2) and Hep-iPS (Hep 1 and 2) cells, as well as 2 human ES cell lines (H1 and H9), analyzed by qrt-pcr in Supplementary Figure 4b were stained for the definitive endoderm markers FOXA2 and SOX17. The number of FOXA2- and SOX17-positive cells relative to the total number of DAPI-positive nuclei were quantified for each cell type analyzed. Values in green, percentage of FOXA2-positive cells; values in red, percentage of SOX17-postiive cells. (b) On d6 of the targeted differentiation protocol, the same cell lines were stained for FOXA2 and the primitive gut marker, HNF1b. The number of FOXA2- and HNF1b-positive cells relative to the total number of DAPI-positive nuclei were quantified for each cell type analyzed. Values in green, percentage of FOXA2-positive cells; values in red, percentage of HNF1b-postiive cells. Scale bars represent 200 μm. 5

Figure S6 Genome browser tracks of the 4 most incompletely silenced genes showing differential methylation between H1 ES cells and IMR90 cells.

6 Figure S6 Genome browser tracks of the 4 most incompletely silenced genes showing differential methylation between H1 ES cells and IMR90 cells. (a) Upper panel, a putative gene C9orf64 on chromosome 9, showing browser tracks for CpG islands, and DNA methylation levels for H1 ES and IMR90 cells assayed by MethylC-Seq (Lister et al. 2009). In the H1 ES and IMR90 cells methylation tracks, the Y-axis displays methylation scores of individual sites (CG, CHG and CHH). Methylation score is defined by the number of Cs and Ts at that position from MethylC-seq reads using the following formula: score = number of Cs / (C+T) * Data from both strands are combined. Scores range between -500 (unmethylated) and 500 (methylated), and the zero line is equivalent to 50% methylated. Negative scores are displayed as green bars and positive scores are displayed as orange bars. Lower panel, a close-up of the promoter near the region (black rectangle) that was assessed for methylation by bisulfite sequencing. (b-d) 3 additional genes which exhibit differential methylation between ES cells and IMR90 cells, and which lack epigenetic reprogramming in ips cells (Supplementary Data 2) (b) CSRP1, (c) TRIM4, (d) COMT. 6

Values were standardized to GAPDH and Ubb. Data are from triplicate reactions. Error bars represent standard deviations.

7 Supplementary Figure 7 a b c Figure S7 DNA methylation and incompletely reprogrammed genes. (a) The high expression level of C9orf64 and TRIM4 in somatic cells and their ips cell counterparts, relative to ES cells (in this case, H9 ES cells), was confirmed by qrt-pcr. Values were standardized to GAPDH and Ubb. Data are from triplicate reactions. Error bars represent standard deviations. (b) The genes which maintain higher expression levels in Hep-iPS cells and Mel-iPS cells compared to ES cells tend to be also methylated at higher levels in H1 ES cells compared to the fibroblast cell line IMR90. The Pearson correlation coefficient between the log expression fold change and single-nucleotide resolution differences in CpG island methylation was 0.37 and 0.74 for Hep-iPS cells and Mel-iPS cells, respectively. (c) We performed additional bisulfite sequence analysis in 4 human ips cells (Hep-iPS 6, Adult Fib-iPS 1, BJ-RiPS 1.2 and BJ-RiPS 1.3), that were derived from different donor somatic cells and by reprogramming strategies different from those that were originally transcriptionally profiled by microarrays. Four additional ES cell lines (HSF6, ESI03, HSF12 and HSF13) were also analyzed, in parallel, as controls. The results are essentially the same as depicted in Figure 4c, while there is some variability in C9orf64 in ES cells, indicating that our findings are not clone- or methodology-dependent. Shown in the graphs is the % methylation observed at the promoters of C9orf64, TRIM4, CSRP1 and COMT in all the original cells that were transcriptionally-profiled by microarray, combined with the results from the newly analyzed ips and ES cell lines. 7

8 Figure S8 C9orf64 overexpression rescues the deficiency in ips colony number that results from C9orf64 inhibition. The number of ips colonies was counted on d21 after infection of BJ foreskin fibroblasts with 4f alone (control), 4f+non-targeting shrna (pgipz-nt) and 4f+C9orf64 shrna (pgipz-c9orf64 i1/2/3). For each condition, C9orf64 was also overexpressed by plove-c9orf64 lentivirus infection, and plove-gfp was used as a negative control. Infections were performed in triplicate, and error bars represent standard deviations. C9orf64 overexpression resulted in a significant rescue of the reduction in number of ips colonies by C9orf64 shrna2 and shrna3, which target the UTR region of the C9orf64 mrna. No rescue was observed for C9orf64 shrna1, which targets the ORF region of the C9orf64 mrna. 8

9 Supplementary Table 1 Gene RefSeq Log 2 (Fib/ES) expression change CpG ES>IMR90 CpG IMR90>ES C9orf64 NM_ TRIM4 NM_ DNAJA4 NM_ IQCA1 NM_ COMT NM_ CES1 NM_ Supplementary Table 1. 6 genes differentially expressed between ESC and all 3 ipsc. DEDS 5% FDR cutoff was used to determine differential expression. 9

10 Supplementary Table 2 Gene name Coordinates(hg18 version) Bisulfite primer names Sequence C9orf64 chr9:85,761,380-85,761,819 Forward TGTAGTTAAGGTAAAGGTTTTTTTT Reverse ACTCAATCCTCAACACCCAAATCTAC CSRP1 chr1:199,742, ,742,959 Forward GTGTTTAGGAAGTTTAGGAAGGTT Reverse CAATATACAAAACCCACTAATTAAC TRIM4 chr7:99,354,907-99,355,386 Forward ATAGTTTAGGTAGATGGGGTAGGTTAATTT Reverse CCTAAACCCTCAAACTTAAAAAAAA COMT chr22:18,309,072-18,309,357 Forward TTTGAGTAAGATTAGATTAAGAGGT Reverse ACAACCCTAACTACCCCAAAAACCC Supplementary Table 2. Sequences for bisulfite primers used for methylation analysis. 10

11 Supplementary Table 3 Gene name Forward primer sequence Reverse primer sequence GAPDH CAATGACCCCTTCATTGACC GACAAGCTTCCCGTTCTCAG Ubb TTGTTGGGTGAGCTTGTTTG GTCTTGCCGGTAAGGGTTTT TBP TGTGCACAGGAGCCAAGAGT ATTTTCTTGCTGCCAGTCTGG C9orf64 AGTGGGTACTGGTCCCTGTG GTCGCGTAGTACGAGGCACT Endogenous OCT4 TGTACTCCTCGGTCCCTTTC TCCAGGTTTTCTTTCCCTAGC Endogenous SOX2 GCTAGTCTCCAAGCGACGAA GCAAGAAGCCTCTCCTTGAA Endogenous cmyc CGGAACTCTTGTGCGTAAGG CTCAGCCAAGGTTGTGAGGT Endogenous KLF4 TATGACCCACACTGCCAGAA TGGGAACTTGACCATGATTG Endogenous NANOG CAGTCTGGACACTGGCTGAA CTCGCTGATTAGGCTCCAAC Lentiviral OCT4 CCCCTGTCTCTGTCACCACT CCACATAGCGTAAAAGGAGCA Lentiviral SOX2 ACACTGCCCCTCTCACACAT CATAGCGTAAAAGGAGCAACA Lentiviral cmyc AAGAGGACTTGTTGCGGAAA TTGTAATCCAGAGGTTGATTATCG Lentiviral KLF4 GACCACCTCGCCTTACACAT CATAGCGTAAAAGGAGCAACA Lentiviral NANOG ACATGCAACCTGAAGACGTG CACATAGCGTAAAAGGAGCAA TRIM4 GAAGTGAAGAACGCCACACA TCAACCAGGAAGTTGTGCAG SOX17 GGCGCAGCAGAATCCAGA CCACGACTTGCCCAGCAT FOXA2 GGGAGCGGTGAAGATGGA TCATGTTGCTCACGGAGGAGTA MSX1 CGAGAGGACCCCGTGGATGCAGAG GGCGGCCATCTTCAGCTTCTCCAG BRACHYURY TGCTTCCCTGAGACCCAGTT GATCACTTCTTTCCTTTGCATCAAG NCAM AGGAGACAGAAACGAAGCCA GGTGTTGGAAATGCTCTGGT SOX1 ATGCACCGCTACGACATGG CTCATGTAGCCCTGCGAGTTG OCT4 T GCATAGTCGCTGCTTGATCG TGGGCTCGAGAACCATGTG CXCR4 CACCGCATCTGGAGAACCA GCCCATTTCCTCGGTGTAGTT SOX7 ACGCCGAGCTCAGCAAGAT TCCACGTACGGCCTCTTCTG HNF1B TCACAGATACCAGCAGCATCAGT GGGCATCACCAGGCTTGTA HNF4A CATGGCCAAGATTGACAACCT TTCCCATATGTTCCTGCATCAG Supplementary Table 3. Sequences for qrt-pcr primers used in our study. *OCT4 T primer set was used only for the targeted differentiation analysis. 11

Xeno-Free Systems for hesc & hipsc. Facilitating the shift from Stem Cell Research to Clinical Applications

Xeno-Free Systems for hesc & hipsc Facilitating the shift from Stem Cell Research to Clinical Applications NutriStem Defined, xeno-free (XF), serum-free media (SFM) specially formulated for growth and